The Great Software Schism

Following Nicholas Petreley's discussion of the GNU GPLv3 debate from one angle, I'd like to look at it from another - that of the cultures of the two groups involved - and what this implies for the future.

The constant bickering between those who talk of "free software" and those who prefer the term "open source" is hard to avoid in the computing world.  You only have to look at discussion threads on sites like Slashdot, Digg, Linux Journal or LWN.net to see the two camps rehearsing familiar themes.  This eternal disagreement can be invigorating entertainment at best, or, if you're not in the mood, enervating in its tediousness.  But this fundamental difference of viewpoint has not really been a serious danger to the overall free software/open source movement.  Until now. 

For with the recent publication of the white paper The Dangers and Problems with GPLv3 - and note that to emphasise its earnestness it is framed like an academic paper complete with alphabetical listing of authors and opening abstract - the background skirmishes threaten to escalate into a full-blown war.

On its own, the paper would have been bad enough.  But matters have been exacerbated by the publication of a poll of top kernel coders.  On a scale from 3 ("I wouldn't want to use v2") to -3 ("I wouldn't want to use v3"), 29 of the top kernel hackers were asked which version of the GNU GPL they would prefer to use.  The average was an unequivocal -2.0: "I think v3 is much worse than v2".  But what makes this "informal" poll all the more damaging is that both it and the white paper were instigated by Linus himself:

The reason the poll and the whitepaper got started was that I've obviously not been all that happy with the GPLv3

...

It wasn't meant to be really "definitive" - the poll was literally meant to get _some_ kind of view into how the top developers feel. I think the end result ended up being more definitive (just thanks to the very clear voting pattern) than we migth have expected.

So this is no mere rumbling in the ranks, but discontent at the very highest level.

Some of the problems outlined in the white paper seem to be down to misunderstandings of what the GNU GPL v3 is trying to achieve. To correct these misunderstandings, the FSF has released its own "clarification."  But even allowing for these, there remains a key issue over which the two sides differ fundamentally in their views.

This concerns the use of DRM - traditionally "Digital Rights Management", but in FSF-speak "Digital Restrictions Management".  For the FSF - and hence the GNU GPLv3 - the matter is simple:

Some computers are designed to deny users access to install or run modified versions of the software inside them. This is fundamentally incompatible with the purpose of the GPL, which is to protect users' freedom to change the software. Therefore, the GPL ensures that the software it covers will not be restricted in this way.

That is, DRM or similar is fundamentally antithetical to users' freedom, and therefore cannot be countenanced.  But the authors of the white paper see things differently:

While we find the use of DRM by media companies in their attempts to reach into user owned devices to control content deeply disturbing, our belief in the essential freedoms of section 3 [of the white paper] forbids us from ever accepting any licence which contains end use restrictions. The existence of DRM abuse is no excuse for curtailing freedoms.

And what exactly is the essential freedom they are most worried about? 

the freedom from binding the end use of the project. Without this freedom, it would be much more difficult to satisfy the objectives of the contributors, since those objectives often have expression in terms of the end use to which they wish to put the particular project. Therefore, in order to maintain the essential development synergy and consequent innovation stream it provides to Linux, we could not countenance any change to the GPL which would jeopardise this fundamental freedom.

That is, it is a freedom not so much for the users, but for the programmers.  The white paper explains why its authors believe this is so important:

individuals with disparate (and sometimes even competing) objectives can still march together a considerable distance to their mutual benefit. This synergy of effort, while not compromising dissimilar aims, is one of the reasons Linux manages to harness the efforts of not only motivated developers but also corporate and commercial interests.

So, according to this view, kernel programmers must have the freedom from end use restrictions on the project or else this "synergy of effort" that has made Linux so successful will be lost.  In other words, at the core of the kernel hackers' argument is a pragmatic concern: GNU GPL v2 is better because it will allow Linux to be more successful.  This emphasis on success is made quite clear from the very first sentence of the white paper:

Over the past decade, the Linux Operating System has shown itself to be far and away the most successful Open Source operating system in history.

Linus, in his self-described "ode to GPLv2", underlines the pragmatic viewpoint with a telling parenthetical remark.  He says:

You can use the end result any way you want (and if you want to use it for "bad" things, be my guest), but we ask the same exact thing of everybody - give your modifications back.

That's true grace. Realizing that the petty concerns don't matter, whether they are money or DRM, or patents, or anything else.

This is the heart of the difference between the free software and open source worlds.  For the latter, GPLv2 is a highly-efficient licence because it works incredibly well as a way of getting the largest number of people working together to write great code, whatever their own particular interests or agendas - that's "true grace".  But for the free software camp, using it for "bad" things is anything but "true grace" - quite the opposite, in fact; for RMS and his followers "petty concerns" like DRM or patents do matter - in fact, they matter far more than whether a project is successful.  They matter because they negate the whole point of the GNU GPL, as stated right at the start of the licence:

the GNU General Public License is intended to guarantee your freedom to share and change free software--to make sure the software is free for all its users.

Or, as RMS said when I interviewed him in 1999:

The overall purpose is to give the users freedom, by giving them free software they can use, and to extend the boundaries of what you can do with entirely free software as far as possible.  Because the idea of GNU is to make it possible for people to do things with their computers without accepting domination of somebody else. Without letting some owner of software say, I won't let you understand how this works, I'm going to keep you helplessly dependent on me and if you share with your friends I'll call you a pirate and put you in jail. I consider that immoral, and I'm working to put an end to that way of life.

There is no room for compromise in this purist world-view.  As RMS emphasised in 1999:

the only reason we have a wholly free operating system is because of the movement that said we want an operating system that's wholly free, not just 90% free.

Hitherto those in the open source world have been able to let RMS live his uncompromising life because his system has worked amazingly well, and because they are pragmatists: they didn't really care about the politics implicit in the GNU GPL provided it didn't get in the way.  But with GPLv3, it does get in the way, and the publication of the white paper and poll seems to mark an end to the truce that has existed between the two sides for nearly a decade.

The white paper ends with a plea:

we implore the FSF to re-examine the consequences of its actions and to abandon the current GPLv3 process before it becomes too late.

But as its authors surely know, there is no hope that the GPLv3 will be abandoned - not after the 16,000 emails or so that RMS and Eben Moglen, the FSF's General Counsel, have exchanged on the subject.  Modified slightly, perhaps, but not in its substance, for example to weaken the clause aimed at DRM schemes: for purists, there can be no dilution of their basic beliefs - no "90%" solutions.  Unfortunately, the pragmatists too seem to be painting themselves into a corner with their definitive rejection of the DRM clauses.

There is no obvious way out of this situation.  The kernel coders will presumably stick with GNU GPLv2, as will others who are focused on programming efficiency and success; meanwhile, the GNU project and those programmers who agree with its unbending ethical viewpoint will adopt GPLv3.  As a result, some code will be forked, with dual GPLv2 and GPLv3 licensing, and there will be growing confusion in the world of code.

The Great Software Schism will have begun.

Glyn Moody writes about free software and open source at opendotdotdot.

______________________

Comments

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.

Facts on the ground

dhlii's picture

Twenty years ago no sane lawyer would argue that a shrink wrap license agreement was valid. No laws have changed - though efforts have been made. Vendors and "infringers" alike are scared to go to court on the matter, but today it is increasingly likely that courts would find a shrink wrap license valid - despite the fact that it flies in the face of fundamental black letter contract law. Judges actively avoid decisions that radically change the status quo. No amount of money and lobbying will persuade congress to pass laws against activities that a substantial minority of people routinely engage in. But it is easy to buy legislation when no one feels threatened. We have had over a quarter of a century of heinous Intellectual Property laws - most ordinary people roll their eyes when talk drifts to IP. The DMCA exists not because people support it, but because it effected very few people and ordinary people did not care.

hoping that the GPLv3 will build sufficient facts on the ground to impede future evil IP laws, is very optomistic. But it is alot more hopeful than expecting congress to do the right thing on their own, without a body of loud and obvious victims.

Besides look where political channels got us on Net Neutrality ?
Not withstanding Micheal Powells admonishion that the worst thing that could happen to Net Neutrality advocates would be getting what they want. The best ordinary people can hope for out of the legislative process most of the time is gridlock.

At least there are some things we agree 100% on!

Freeman's picture

I agree 100%. But first, I think it is a non-sequitur to say that because TiVo does X, it follows that it will lead to Y, which is a locked-out general purpose PC. There's no connection.

No, what I said was There seems to be an industry thrust in that direction (Trusted Computing, etc.). I'm saying Y is underway, not necessarily connected to X other than X provides a proof-of-concept model and exposes a vulnerability in the GPL.

One major reason network computing never took off, despite the fact that they would reduce TOC for end-users (esp. corporations), is because hardware companies don't want to sell them!
...
The market will avoid a locked-down PC.

You mean the same market that can't avoid increased TCO because hardware companies only sell what they want to?

I think the only way a locked-down PC could succed is if it were forced on us by some ridiculous law, like an offshoot of some DRM law like DMCA.

Sure, but it could never happen AGAIN! The government could never be persuaded to prevent us from modifying our software in order to protect us from hackers and terrorists, and make it easier for them to spy on the terrorists in our midsts, right?

In that case, we should be lobbying our representatives to get rid of such laws.

Because that worked so well last time. DMCA sure was a close call, though, wasn't it?

I think if Linux became GPLv3, it wouldn't affect hardware design in the least. It would simply force hardware makers who want to avoid the hassle of GPLv3 to adopt something other than Linux. There are plenty of good alternatives, like *BSD,

It's not about hardware design. You said If it ran a proprietary OS, we wouldn't even be having this discussion. On that we agree 100% (wow, twice in one day -- we're making progress!). In fact, I wouldn't object if Tivo used a BSD kernel and kept their own modifications secret, because that license would permit it (in spirit as well as letter), It's not the hardware that is objectionable, it is the fact that Tivo reserves for itself more rights to the GPL software in the product that they distribute to their customers than those customers are granted, a clear violation of the spirit of the GPL, if not the letter, by exploitation of a legal loophole.

Like I said in this reply to Anonymous, Karl O. Pinc said it best (he was replying to you): The one and only intention of the GPL is to give the recipient of software rights to the software that are identical to the author's/distributor's rights. TiVo has the right to run modified software on the hardware they sold you, you do not. Hence TiVo has violated the spirit of the GPL and the GPL is being patched so that this problem cannot reoccur.

Ponder this question: What would you say if you bought a new Tovi brand computer with GNU/Linux pre-installed and found you could not modify the software that came on your computer or add new software to expand it's functionality? (It wouldn't necessarily have to be hardware that locks you out) Would you say "That's alright, I can just go buy some new open hardware, assemble it myself and compile all this source code Tovi made available and then I'll be able to modify and update the system at will"?

So the right approach to

Rufus Polson's picture

So the right approach to software licensing is . . .
to trust the market?
Yeah, like "the market" got us Free Software in the first place. If you'll recall, "the market" got us Microsoft and a monopoly. This seems an ill-conceived strategy.

Huh?

Nicholas Petreley's picture

If the market didn't buy into Linux, then how did it get its current market share? Yes, the market DID make free software successful. And I'd like to think that I'm part of the reason, since I was singing the virtues of Linux back in 1995 in InfoWorld when practically nobody else in the corporate market even knew what it was.

If it wasn't for the market, Linux would still be something only hobbyists and hackers use. And I'm betting that if Linux ever adopts GPLv3, it'll go back to being an OS that only hobbyists and hackers use.

Huh??

dhlii's picture

The window of opportunity for commercially co-opting the GPLv2 is extremely narrow. In all probability Tivo is the first serious vendor to attempt such a serious violation of the spirit of the GPL. Why are vendors that already accept (as they decry them) the limitations of the GPLv2, likely to suddenly change their minds when a loophole most either were unaware of or would not dare try to pass through is closed ?

I expect that the adoption of the GPLv3 will create continuous problems for the kernel developers until they eventually atleast allow GPLv3 contributions.

I expect that a very small number of business will jump ship - is anyone expecting hollywood and the entertainment industry to jump onto the GPL/Linux bandwagon anytime soon ?.

Otherwise little will change.

If anything

cprise's picture

...sampling your articles here, I'd say that you've probably contributed to Linux becoming a balkanized misnomer on the desktop. Deceiving people (even by oversight) into thinking of distros as "Linux" has hurt the acceptance of FOSS operating systems immeasurably. "Is this application Linux compatible" is a common yet inappropriate question in this turn of events, because people like you persistently imply that fleshed-out distros define a real PC platform having the identity "Linux". The meme has afflicted too many geeks with a prideful schizophrenia.

Only hobbyists and hackers and sysadmins use Linux, because Linux cannot possibly be meaningful to end-users. We can see and touch the kernel; they cannot. When we do not realize that users cannot see/touch the real Linux, then we have become ignorant.

The raw installed base and public mindshare of "Linux" remain tiny. But hey, at least sysadmins are getting a taste of liberation, even if Linux becomes next decade's Netware (in that case I'm sure expressions of fondness for the old penguin will get embarrassing at times). To spell it out for you: Being hip for a while with sysadmins did not save Netware.

Where is the real success, especially with .NET services taking off like a rocket this year (at the expense of FOSS)?

Wait, nevermid. I remember now: Linux is supposed to be reserved for the elite.

Re: Huh?

Anonymous's picture

True, but at least they'll feel good about it :)

Well put

Glyn Moody's picture

Yes, this is exactly the point I was making, and your comparison with BSD vs. GPL is well made.

Allow software developers to disable DRM in GPLv3

Vass Zoltan's picture

Software developers should be allowed to choose if their GPLv3 software
is under the DRM related section or not.
Example:
"This Software is under GPLv3 exept sections regarding DRM"

You can't.

wazoox's picture

Check FSF website. You simply aren't allowed to call a licence GPL if you modify it in any way. Take all or nothing... See
GPL FAQ

Thank You!!!

Freeman's picture

Glyn,

Thank you for an extremely well-written article that explores the issue honestly and provides some very valuable insights into the views on both sides of the issue. I couldn't have done as well myself, but I would like to emphasize the point that the conflict seems to be rooted in the two sides' very different definitions of "success".

Your conclusion highlights just how cleverly this rift between the two camps has been exploited with the DRM strategy, which now appears to possibly have a "divide and conquer" component to it.

I was just about to unsubscribe from Linux Journal after the invective-filled, name-calling, strawman-propping treatment I received while discussing Nicholas Petreley's article when I merely tried describe why I can understand the FSF's desire to update the GPL and to pass that understanding on to others (something you were able to do far better than I), but you have redeemed my perception of the quality of this publication with your article, and then some.

From now on, I'll be looking forward to reading your articles. As for Petreley, well, I can't say that I consider the accuracy of his writing up the standards you have acheived with this one, but I'm going to go ahead and have a look at the "Pet Peevo" article anyway. I just doubt I'll feel like joining the discussion over there, as I've seen more well-reasoned arguments over on Slashdot at the -1 mod level.

And thank you....

Glyn Moody's picture

...for the kind words.

I think it's important to accept the passion that this subject generates as part of the power that drives free software and open source. Yes, it can lead to some hasty and sometime hurtful words, but it's a signal that people care, and profoundly. In an age where so many seem completely apathetic and disengaged, this is something to be grateful for. The trick is to retain a broader perspective on things, which is what I was attempting to contribute to, as even-handedly as I could.

Software is a tool

Joe Klemmer's picture

It is a tool to be used to perform some task or accomplish some requirement. It's a frelling screwdriver. It is not the Declaration of Independence or the Magna Carta. The problem with the whole FSF/GNU/GPL thing is the bastardization of the word "freedom" that one side of this debate has performed. Here's the definition of the word -

freedom
n 1: the condition of being free; the power to act or speak or think without externally imposed restraints

I have emphasized the relevant portion of the definition for clarity. Anything, and I do mean ANYTHING that curtails, limits, restricts or inhibits is, by definition, not free.

Now, don't get me wrong. I am a firm backer of the GPL (v2) and I hold a lot of what the FSF is fighting for to be good and proper things. It's just that they are not fighting for freedom. They are fighting to impose specific limitations on software. I agree with the limitations expressed in GPL and the LGPL. They are good and proper things. But the biggest mistake that RMS ever made was to associate his philosophy with the word "freedom."

Software is, as I stated, a tool. It exists to be used. How it is used is, and should always be, immaterial to the software itself. There are external laws and methods to handle the improper use of things that are built by tools (including software).

I have stated this position in the past and the rebuttal is always the same. Lots of words and rhetoric with no concrete, explicit, factual, rational content. I expect more of the same in response to this comment.

--
Indie Game Dev and Linux User
Contact Info: http://about.me/joeklemmer
"Running Linux since 1991"

Freedom for everyone

Anonymous's picture

Freedom without limitation does not work well!
In democratic societies we impose limits on ourselves, such that all of us can have the SAME freedom.
Complete freedom would allow me to kill a neighbor but we don't allow that because it limits my neighbors freedom.
Democratic freedom: I'm allowed to do anything as long as it doesn't prevent someone else to do the same.

This is the basic law of democracies as well as the principle behind GPL.

Now we have a loophole in GPL that needs fixing, hence we're working on GPLv3. Very basic, very simple and it avoids petty concerns about money etc.

Print is a tool

Rufus Polson's picture

Print is a tool. Printing presses are pieces of hardware. They aren't the declaration of independence or the Magna Carta. I don't understand what all this nonsense about freedom of the press is about . . .

free software is not about ANY freedom

Anonymous's picture

It's just that they are not fighting for freedom.

that is correct. free software is not about fighting for freedom in general. it never was. it is fighting for very specific types of freedom. which types of freedom these are is very clearly described in the license.

to defend these freedoms others must be restricted. just like the right to live requires us to restrict the right to kill someone.

the defintion of freedom you give is correct, but not practically implementable, because it would be equal to anarchy.

greetings, eMBee.

re: Software is a tool

Freeman's picture

Joe,

While I respect your well-articulated viewpoint, I worry that software "freedom" by any definition of the word will no longer exist in any meaningful form once DRM becomes commonplace.

I don't have "Lots of words and rhetoric" for you, just one question: The FSF has made a strenuous and valiant attempt to deal with the DRM issue with regards to the future of (shall we call it for the sake of argument) free and/or open-source software development. Who, besides the FSF, has publicly offered any proposal to counter the threat to F/OSS development posed by DRM?

Criticism is pointless if no alternative is proposed.

Not from me

Nicholas Petreley's picture

I have stated this position in the past and the rebuttal is always the same. Lots of words and rhetoric with no concrete, explicit, factual, rational content. I expect more of the same in response to this comment.

You won't get it from me. I agree with you 100%. This is where I part ways with GPLv3. It imposes a philosophy on how people must use (or not use) GPLv3 software that is unrelated to the software itself. We can argue all day about the good or evils of DRM, DMCA, and so on, but that's a different topic. The FSF is trying to use Linux as a weapon against these things, and I'm glad Linus and others are not cooperating. If the FSF wants to fight that battle, they should do it some other way, not by trying to wield Linux as their sword.

Why would FSF need to "wield Linux"?

Freeman's picture

Hate to break it to you, but Linux ain't the whole enchilada, it's just the kernel. A large part of any GNU/Linux distribution is the GNU part. Which version of the GNU Public License do you suppose the GNU organization will adopt?

How can you say that use is unrelated to the software?

Karl O. Pinc's picture

If the software is GPL v2-ed and I can modifiy it all I want, but I can't actually run it because of DRM, then what good is my ability to modify the software? I don't understand how you can say that the use of the software is unrelated to the software itself. Software that can't be used may as well be literature, it's not software at all any more.

Sigh

Nicholas Petreley's picture

For the umpteenth time...

Can you modify the TiVo source code en masse to run on your own hardware? Yes. Can you use (and therefore benefit from) the modifications TiVo made to the software? Yes. You can incorporate some of their goodies into your own project, for example. TiVo has complied with the letter and spirit of the GPL. It has made its modifications available. It has given back to the community the value it has added to the code. It offers it as a free download, no less, (unless you want it on a CD).

Just because your modifications won't run on a particular box -- one you don't even have to buy -- doesn't make the code useless. Just because it might be difficult to extract the good stuff for your own purposes to run on a different machine doesn't make it useless.

Now we may be getting somewhere

Karl O. Pinc's picture

Please see my comment above entitled "Execute on something useful".
We may have gotten to the core of the argument. (I love it when that happens. Everybody knows exactly what they're agreeing to disagree on. :)

There's nothing to comment about

Nicholas Petreley's picture

Your premise is flawed. You do not have to buy a TiVo to get the software. You get the executable binary when you buy a TiVo. The GPL prevents people from keeping their modifictions to GPL code private. TiVo has made their modificationis publicly available.

Down to the crux then

Karl O. Pinc's picture

The GPL does not require TiVo to make the source public, they chose to do so.

It seems then that our fundamental disagreement is the intention of the GPL. If we disagree about this then there really is nothing more to say.

The one and only intention of the GPL is to give the recipient of software rights to the software that are identical to the author's/distributor's rights. TiVo has the right to run modified software on the hardware they sold you, you do not. Hence TiVo has violated the spirit of the GPL and the GPL is being patched so that this problem cannot reoccur. The reason why Tivoization is discussed so much is because it's the most obvious exploitable bug in the GPL v2, and because TiVo has demonstrated to the world the potential profit in exploiting it.

Various people, perhaps including Linus, may have chosen to license their software under the GPL v2 for reasons only peripherally related to the purpose of the GPL, but the GPL and the FSF have _always_ been clear as to their reasons for creating and using the GPL. Nobody using GPL v2 is forced to switch to v3, but nobody should be suprised either when they find that the v3 does a better job than the v2 when it comes to enforcing the end-user's rights. Which is the point of the article.

It's no wonder that people are suprised regardless. When I first encounted it I hated the GPL. This persisted for years until I began to pay attention. At that point it took 2 years and numerous re-readings to understand it, it's implications, and everything it does for me as a businessman, developer, and end-user. The "impractical" "pie in the sky" ideals of the GPL turn out to have a lot of knock-on peripheral benefits that people seem to think they can get without having to enforce the ideals. This has turned out not to be true in the past and I believe it won't be true in the future either, but time will tell. It took ~20 years to find signficant bugs in GPL v2 and just because a market has grown up surrounding those bugs is no reason not to fix them. In fact it's every reason to fix them. By showing consistency of purpose the FSF can only strengthen confidence in the Free Software market and enhance it's long term prospects.

OMG

Ookaze's picture

Can you modify the TiVo source code en masse to run on your own hardware? Yes

No you can't. Stop saying lies like that. If I bought a Tivo (my own hardware) and modified the Tivo source code, I couldn't run it on the Tivo (my own hardware).
So you're wrong already, and so the spirit of the GPL is not respected at all.
What's this denying of reality saying that this respects the spirit of the GPL ? If that was true, why would there be a problem between the Tivo and GPLv3 ?
This is very basic logic indeed, I'm amazed you manage to then say Tivo respects the spirit of the GPL.

Your own hardware

Nicholas Petreley's picture

By "your own hardware" I meant "hardware you designed yourself" as in a PC with your own choice of capture card, etc.

I obviously have to start being ridiculously explicit about "your" and "their" because people keep interpreting that to mean whether or not TiVo owns its box after you buy it, when that's not what I'm talking about at all.

You can't keep that simple concept straight

cprise's picture

...because doing so results in your arguments collapsing.

The only way to make your position (Tivo "owning" something in my home) work is to support the new and radical concept of "intellectual property" that Wall St. is in love with.

Please do us a favor: Try not to be so petulantly dismissive of key concepts like real property, because it makes the rest of us sad to even look at LJ.

Not so much tool as innocent bystander

Glyn Moody's picture

I don't think the FSF is trying to use Linux or GNU/Linux as a tool against these things: it is trying to fight them through the licence. And the licence was drawn up by Stallman with the explicit intention of fighting those things - not of creating a great operating system, which was a means not an end. GNU/Linux is just an innocent bystander caught up in this because it happens to use the GPL.

The point I was trying to get across in the article was that the two sides of this argument are talking at cross-purposes: in a sense they are both right and both wrong. For Stallman, keeping the code free of external constraints is paramount: the constraints the licence imposes are only there to achieve that (the paradoxical nature of this is something that surely tickles RMS's hacker nature). He doesn't care whether the applications of the licence - like GNU/Linux - would be more "successful" without the DRM provisions: he would argue that the success is illusory because it's not propagating true freedom of the kind he strives for.

However, for the pragmatists like Linus, this is clearly nuts, because it threatens the uptake of projects like GNU/Linux, and so undermines its own success. But again, as I wrote in the piece, for the purists, it is better to fail without compromise, than succeed by compromising. The trouble is, there is no possible common ground between these two viewpoints, which is why I see some kind of split looming for the free software and open source community.

Clearly nuts . . . .

Anonymous's picture

Glyn,

Good article, well balanced and a fair description of the problem - as far as it goes.

But for the free software camp, using it for "bad" things is anything but "true grace" - quite the opposite, in fact; for RMS and his followers "petty concerns" like DRM or patents do matter - in fact, they matter far more than whether a project is successful. They matter because they negate the whole point of the GNU GPL, as stated right at the start of the licence . . .

I understand the point you're making here (on behalf of RMS, the FSF, and your first commentator Freeman), and I agree with this philosophy whole-heartedly.

. . . for the pragmatists like Linus, this is clearly nuts, because it threatens the uptake of projects like GNU/Linux, and so undermines its own success.

I can see this point as well, I believe that the current form of the GPLv3 will cripple corporate adoption of any "free software" - Linux or otherwise. Further, I don't believe it takes a brain surgeon to figure out why. Thus it becomes very clear that there will be a parting of the ways between the "free software" and the "open source" camps if things progress as they have to this point.

I did read (and participate in) the discussion that Freeman mentioned above, and I understand the passion on both sides of the debate. Unfortunately, we will likely spend our time attacking each other rather than the more serious issues (DMCA, DRM abuse, "trusted computing" etc.) which we share a common disdain for.

Lets take a second and cut straight to the core issue, which I don't believe has received the coverage it should: For Stallman, keeping the code free of external constraints is paramount: the constraints the licence imposes are only there to achieve that . . . .

And thus, RMS, the FSF, and Freeman above become exactly the thing they hate. I'll ask the same question I've asked in the past: how is the core intent of the GPLv3 any different than DRM abuse? Quoted from my previous post on the same subject: What you're saying is that "freedom isn't free unless I agree with how it is used, otherwise develop your own solution - I won't let you freely use my free code." The GPLv3 and DRM abuse are merely two sides of the same coin, albeit one for "good intentions" and one for a profit motive. But here's the big question: Who appointed RMS, the FSF, or Freeman as moral arbiters of what's a right and proper use of their freely contributed code? As I've said, it's really easy to become the very thing you hate, when you loose your perspective.

As someone mentioned in Nicholas' previous post, attempting to end war, hunger, and unfair control are admirable goals, things we should all work towards. But they have very little to do with a software license. And imposing unfair control in a software licence to try and end other acts of unfair control is simply stupid.

Solution? I wish I had one; for that matter I wish RMS and the FSF had one. The GPLv3 will hurt far more than it will help, and potentially set the "open source" movement back years. It will probably kill the "free software" movement completely.

Nicholas mentioned the idea of a "free hardware foundation." Couldn't hurt, and I suspect it'd have a MUCH better chance of industry adoption than the zealotry currently championed by the FSF.

I can see this point as

Ookaze's picture

I can see this point as well, I believe that the current form of the GPLv3 will cripple corporate adoption of any "free software" - Linux or otherwise

Why ? If you believe that, then it supports the GPLv3 provisions.
Because the corporates that would not adopt GPLv3 software are all those that want to do the same thing as Tivo.
Which would be a very BAD thing for free software.
So basically, people that fear GPLv3 because of DRM, would better live in a world where their code is mostly used in Tivo-like hardware ?
So as soon as DRM is ubiquitous on PC, they will have sowed their own demise, unable to run their own code on anything anymore.
This is plain stupid and shows a very short term centric view. But Linus isn't known for his wisdom about licenses and middle term view (like BitKeeper showed us).
I mean, this DRM clause should be NO problem at all, unless the people using the code actually want to lock it behind keys.

Who appointed RMS, the FSF, or Freeman as moral arbiters of what's a right and proper use of their freely contributed code?

The people who chose their license. Next question.

As someone mentioned in Nicholas' previous post, attempting to end war, hunger, and unfair control are admirable goals, things we should all work towards. But they have very little to do with a software license

That's why the software license addresses only the use of the source code. For example, it wouldn't force Tivo to design their hardware any way they don't like, they'd just not be able to use the GPLv3 code if not compliant.

And imposing unfair control in a software licence to try and end other acts of unfair control is simply stupid

It's so stupid that people like you think it matters to talk about it. Perhaps it's not so stupid after all.
I wonder what is unfair too. If I don't want my software locked in Tivo and use GPLv3 on my code, how is it unfair ? Do I owe Tivo any code ?

Solution? I wish I had one; for that matter I wish RMS and the FSF had one. The GPLv3 will hurt far more than it will help, and potentially set the "open source" movement back years. It will probably kill the "free software" movement completely

The very same argument was made until not so long ago for GPLv2. Funny !
You want to impose the Open Source viewpoint on the Free Software community, like was done before.
Of course, you have to acknowledge that there are more GPL code than any other licensed code, so you fear this change.
If the Free Software is killed by this, it means it couldn't have succeeded, and was bound to fail ultimately, because it was used as free labor.
Like the article say rightly, lots of people just shrugged off the Free Software proponents, because they could avoid the political things.
Now they can't, and will have to live with the reality of what FSF is.

How is designing a license and letting you have a say in it, zealotry ? You're not even forced to use it.
You sound like the zealot actually.

Excellent!

Freeman's picture

Excellent response. Sometimes I think the "open-source movement" should just adopt the BSD license and be done with it, especially since they seem to either not understand or ignore the purpose of the license. Who knows, maybe GPLv3 will push things in that direction, for better or worse.

Of course, GNU/Linux succeeded to a far greater degree than any of the BSD's, and most people attribute that to the choice of the GPL for the license, so a switch to BSD might kill the golden goose.

Maybe they should spend the time and effort to develop their own license (OSL?) if they don't like the ones freely available from the FSF and elsewhere.

re: clearly nuts

Freeman's picture

I did read (and participate in) the discussion that Freeman mentioned above, and I understand the passion on both sides of the debate.

hehe, it did get a little nuts over there, didn't it? I figured next we'd get teenage girls arguing over who's better looking -- Richard Stallman or Eric Raymond!

And thus, RMS, the FSF, and Freeman above become exactly the thing they hate. I'll ask the same question I've asked in the past: how is the core intent of the GPLv3 any different than DRM abuse?

Your point about becoming what you hate is well-taken, it's all too common. But consider this: the same argument could be made for the original GPL. How is the core intent of the GPL any different than copyright restrictions? RMS designed the GPL to use copyright restrictions to enforce the right to freely modify, share, and use downstream all the way to the last user. That has always been the intent of every version of the GPL. You "become what you hate" when you abandon your core values and take on those of the enemy in order to fight them. I don't see RMS doing that any time soon, but if, like you say, he has, then it happened a long time ago.

Glad to see you stuck

Anonymous's picture

Glad to see you stuck around. While I firmly believe the "there's no right answer" philosophy is total BS, I also believe that sometimes getting to the "right answer" takes time, work, patience and cooperation. I firmly hope that the open source and the free software camps can figure that out, before doing any permanent damage to each other.

On the subject of "the right answer", as I've said in the past, I wholeheartedly agree with the "philosophy and spirit" behind the GPLv3, and I wish the FSF every success. I work hard to educate people on the evils of DRM abuse (thank you Nicholas for correcting my use of the "DRM" term), and the sickening mistake that is the DMCA.

In a perfect world, we own the hardware we buy, and we have every right to do what we wish with open source, or free software code. In this perfect world, hacking hardware or software we own to operate to our specifications is permitted - and even encouraged as long as we continue to "give back" to the community.

We don't live in a perfect world I'm afraid, and thus the results will be as pointed out by Nicholas farther down this thread. The GNU Hurd kernel will be released under what will become the GPLv3, which is a grand philisophical statement, but used by no one - corporate, home, or otherwise.

The "threat of switching to the GPLv3" will worry no one. Current Linux code is under GPLv2, and can be used even if future versions migrate to GPLv3. We would simply see the same fragmentation that doomed the Unix market. Again, bad for us (the users).

I believe you're correct about RMS. He means well. And while I share many of the same "core values," as the FSF, I still believe that their methodology leaves a lot to be desired. It seems that RMS believes "the ends justify the means." And in this sense, I do believe they have become what they hate most; they may have kept their "core values", but their methods have become decisively more draconian. That was not true with the GPLv2, where the focus was on the availibility of the code, not the proper use of that code - and THAT is the primary reason the GLPv2 has been such a grand success.

In any case, as I said, I'm glad to still have your participation in this discussion - as well as the other rational readers here. Hopefully Linus and RMS are paying attention to the larger issues, outside of their own sphere of interest.

Sticking around

Freeman's picture

Thanks for your encouragement, and it's good to see you over here too, even though we don't see eye-to-eye on everything.

We don't live in a perfect world I'm afraid, and thus the results will be as pointed out by Nicholas farther down this thread. The GNU Hurd kernel will be released under what will become the GPLv3, which is a grand philisophical statement, but used by no one - corporate, home, or otherwise.

I can't say I agree with this. While few people may be using Hurd, GNU isn't just the Hurd kernel, and GNU/Linux isn't just the Linux kernel. There are the GNU tools to consider as well, which we can also expect to migrate to GPLv3. These tools are a vital part of any GNU/Linux distribution. Anyone hoping to use GNU/Linux who are unwilling to abide by the GPLv3 will have to find replacements for the GNU parts or maintain the GPLv2-licensed versions themselves (we can confidently predict forking here). While this may not affect Tivo too much (they probably are more interested in the kernel for their purposes -- Do you need to run Emacs on a Tivo? etc.), I expect it would affect general-purpose computing a lot.

I regret that I allowed myself to get so bogged down in the Tivo-specific hardware arguments in response to Nicholas' article, because the tivoisation of general-purpose computing is the real issue. I don't really care that much about Tivo, I even agreed with Nicholas early on that since I don't like their lock-down, I'm not buying one. But if we stand by and do nothing while "trusted computing", etc. locks down our general purpose computers and their peripherals, how will that affect free / open-source software development?

The "threat of switching to the GPLv3" will worry no one. Current Linux code is under GPLv2, and can be used even if future versions migrate to GPLv3. We would simply see the same fragmentation that doomed the Unix market. Again, bad for us (the users).

This one I can agree with. There is a real danger if we don't find a way to mitigate the licensing issues. GNU/Linux already has a significant amount of fragmentation. Just as an example, recently I had to compile something for a Mandriva box because the only pre-compiled version I could find was a .deb -- no problem for me, but for average-joe end user it's a different story.

And in this sense, I do believe they have become what they hate most; they may have kept their "core values", but their methods have become decisively more draconian.

Can't disagree with the draconian part. Of course, the threat they're fighting has used increaingly draconian measures as well. Like I inferred before, maybe an argument could be made that the GPL sold out from the very beginning by resorting to the same methods as those used by proprietary software companies. On the other hand, I haven't seen any other organization propose a realistic plan for countering DRM abuse. Certainly you don't think we should just stand by and watch them lock us out of our own computers, do you?

Lockout

Anonymous's picture

Freeman wrote:

Certainly you don't think we should just stand by and watch them lock us out of our own computers, do you?

Absolutely not. It seems we agree 100% on the core issues here. It scares me to death to think that the DMCA will stand, and that the "treacherous computing" issue will blosom into the horror that it could very well become. I simply don't believe that resorting to the same methods used by big brother, big media, or Micro$oft is the way to beat them.

For a solution, I'd look at the strengths of the GPLv2, and the progress of the OSS movement:

1) Freedom to use the code as you see fit.
2) Wide adoption from the technical community, as an option to tightly controlled, proprietary solutions.
3) Community involvement and contribution, leading to more stable, secure software.
4) Eventual Corporate involvement, using the efforts of FOSS developers to leverage their solutions.
5) Corporate adoption brings funding and resources. Eventually resulting in more features and wider industry (and home) adoption.
6) The product is still OPEN SOURCED. This means that after all the momentum, improvement, and progress, if something takes a turn for the worse (like Tivo's lockdown business model) we're still free to use THE CODE as we see fit, including all their improvements.

Tivo's box is a red herring, it's not really about their hardware. When they use the code, they retain the freedoms guaranteed in the GPLv2. When we use THEIR BOX, we aren't guaranteed those same freedoms to THEIR BOX. But when we use THEIR GPL'ed CODE, we are guaranteed those same freedoms for THAT CODE. But we haven't automatically inherited any extra rights to THEIR BOX. And I believe it's wrong of the GPLv3 to try to dictate that in any manner.

So on the subject of a solution; if someone smarter than myself could come up with a HARDWARE model that matched what was done with the GPLv2 for software, I belive a similar trend would develop (over the next several years) in the hardware and device industry. Consumer education also must play a large part, and the passion you've shown for our freedom is vital in accomplishing these goals.

But I'm afraid I still belive the proposed GPLv3 will do more harm than good. As I mentioned before: even with lofty goals, when you adopt the methods of your enemy, you become the thing you hate.

In any case, I've learned a lot discussing this with you and the other members on the board. Keep fighting the good fight!

free software depends on open hardware

Freeman's picture

So on the subject of a solution; if someone smarter than myself could come up with a HARDWARE model that matched what was done with the GPLv2 for software...

Clearly we agree that free software depends on open hardware. There are attempts underway to put locks on computing hardware. The current popularity of free software may or may not be enough to ensure the availability of open hardware in the future. The hardware model you mentioned is one possible way of providing that availability, and I think I've read about similar efforts underway, but it's certainly well under-the-radar at this point.

Tivo's box is a red herring, it's not really about their hardware. When they use the code, they retain the freedoms guaranteed in the GPLv2. When we use THEIR BOX, we aren't guaranteed those same freedoms to THEIR BOX. But when we use THEIR GPL'ed CODE, we are guaranteed those same freedoms for THAT CODE. But we haven't automatically inherited any extra rights to THEIR BOX. And I believe it's wrong of the GPLv3 to try to dictate that in any manner.

Karl O. Pinc really gets to the crux of the matter:
The one and only intention of the GPL is to give the recipient of software rights to the software that are identical to the author's/distributor's rights. TiVo has the right to run modified software on the hardware they sold you, you do not. Hence TiVo has violated the spirit of the GPL and the GPL is being patched so that this problem cannot reoccur. The reason why Tivoization is discussed so much is because it's the most obvious exploitable bug in the GPL v2, and because TiVo has demonstrated to the world the potential profit in exploiting it.
...
It took ~20 years to find signficant bugs in GPL v2 and just because a market has grown up surrounding those bugs is no reason not to fix them. In fact it's every reason to fix them. By showing consistency of purpose the FSF can only strengthen confidence in the Free Software market and enhance it's long term prospects.

I think your statement "When we use THEIR BOX, we aren't guaranteed those same freedoms to THEIR BOX." should read "When we use the GPL SOFTWARE that came in our Tivo box, we aren't receiving the freedoms guaranteed by the GPL for THAT SOFTWARE". You and Nicholas keep making the point that we have the rights guaranteed by the GPL if we can figure out a way to run Tivo software on some imaginary open hardware, but when we run the code on the box it came in, we don't have those rights.

What would you say if you bought a new computer with GNU/Linux pre-installed and found you could not modify the software that came on your computer or add new software to expand it's functionality? Would you say "That's alright, I can just go buy some new open hardware and compile all this source code my computer manufacturer made available to me and then I'll be able to modify and update the system at will"? How is Tivo different?

Open vs. Closed systems

Anonymous's picture

I'm not sure that significant bugs have been found in the GPLv2. I think it holds up extremely well, and covers everything a software licence has any business covering. As I've said in the past, enforcing the open distribution of code is ethical and the perfect demonstration of the FSF's "consistency of purpose."

I do believe attempting to force closed hardware devices to conform to those same rules simply because they use open code - that they have lawfully given back to the community - is in and of itself an unethical and unjustified attempt to control someone else's private system. As I said earlier, I don't believe the ends justify the means, and RMS and the FSF in this subject are no better in their means than the DMCA.

What would you say if you bought a new computer with GNU/Linux pre-installed and found you could not modify the software that came on your computer or add new software to expand it's functionality? Would you say "That's alright, I can just go buy some new open hardware and compile all this source code my computer manufacturer made available to me and then I'll be able to modify and update the system at will"? How is Tivo different?

Never gonna happen, and I'll tell you why. As Nicholas and others in this thread have mentioned, any computer I buy (or build) will be an open system, on which I will be able to do what I wish. If I had purchased this imaginary closed system you mention, I would have known the limitations of it from day one, and would never have expected support in circumventing this imaginary company's business model or support in overcoming this imaginary device's limitations.

That doesn't mean I wouldn't try - much like I intend to overclock my new Core 2 Duo. But that doesn't mean I expect support to do it, or warranty coverage from Intel, or sympathy if I do something terminal to it - because it's a closed system.

Now I know where anyone reading this will likely go next: the "treacherous computing initiative" will destroy my ability to run GPL'ed code, etc, etc. Yep, you'd be right to say that. And that's where we come back to our common ground. The DMCA must be brought down, the "Treacherous Computing" crap must be stopped, corporations and governments must be made to see that open systems, and open source are the only truely secure systems.

I just don't believe it's a battle the FSF can win with force. The commercial and political world hold far too much influence for that to have any measure of success. Fighting the entrenched powers (and egos) that be with empty threats will simply make the situation worse, and result in the fragmentation and loss I mentioned in a previous post. Thus, the ideals - at some point - still have to function within the practical world, or they are of little value.

The trick is getting to that point of convergence. Something I believe the GPLv2 has done tremendously well in the software world.

Back it up!

Freeman's picture

I do believe attempting to force closed hardware devices to conform to those same rules simply because they use open code - that they have lawfully given back to the community - is in and of itself an unethical and unjustified attempt to control someone else's private system.

First, if you're going to accuse RMS and the FSF of "unethical and unjustified" behavior, back it up with facts like I did here.

Second, you're confusing Free Software with whatever it is you're referring to as "open code". Linus chose to use a Free Software license (GPL) for his code, of his own free will. This Free Software license states that it exists to insure that end users have the right to modify, execute, and distribute the code in addition to the requirement that source code to any modifications must be made available. Now it looks like Linus wants to allow his software to be distributed under non-free conditions, but that is incompatible with the license he chose. It's HIS responsibility to deal with that, not the FSF, who have always been very clear about the purpose of their licenses. That said, I expect he'll just keep doing what he's always done: distribute under GPLv2 and ignore the parts he doesn't like and allow redistributors to ignore it as well. It's up to him to prosecute anybody he finds violating the terms of his chosen license in ways he cares about.

Once again, you've got things backwards: Tivo is trying to force Free Software to conform to non-free conditions simply because they use closed hardware. What's "unethical and unjustified" is Tivo using any method to distribute Free Software under non-free conditions.

There's no problem with hardware manufacturers wanting to prevent modification of any software that might be distributed inside the device, there's just a problem with them using Free Software for that purpose, as it violates the conditions for distibution of that software. This isn't even a big problem for Tivo, because they've already made their own GPLv2 fork of the kernel and can continue using that as long as nobody successfully prosecutes them for violating the stated purpose of GPLv2 (It would sure be interesting to see a (shudder) DMCA prosecution under the "circumvention of copy-protection" clause, because that is exactly what they've done -- see link above), even if Linus decides to migrate to GPLv3. Not only that, but Tivo could satisfy the requirements of GPLv3 by shipping a private key with each Tivo (they don't all have to use the same private key, it's possible to tie a key to an individual box so that the customer can only modify the software in his own box). 99.99% of their customers wouldn't have a clue what to do with it, the rest of us can have our fun (like Linksys WRT54G firmware). If they don't want any of us to have our fun, they really should distribute non-free software in the device -- that would be the "ethical and justifiable" thing to do.

I can't believe how hard it is to get this point across!

Here's why

Nicholas Petreley's picture

Second, you're confusing Free Software with whatever it is you're referring to as "open code". Linus chose to use a Free Software license (GPL) for his code, of his own free will. This Free Software license states that it exists to insure that end users have the right to modify, execute, and distribute the code in addition to the requirement that source code to any modifications must be made available.

That's exactly how Linux works today.

Now it looks like Linus wants to allow his software to be distributed under non-free conditions, but that is incompatible with the license he chose.

Now you're applying YOUR PERSONAL definition of "free" and "non-free" to what Linus is doing. The code is still as free as it ever was. It still fits the GPLv2 just fine. It just isn't free the way YOU want it to be. You believe that free implies the ability to modify the code and run it on specific hardware. But you have yet to cite a single line of the GPL that says anything of the sort.

Rather, the GPL is summed up rather well in the first citation of yours, above. The error in your logic starts when you add your own personal interpretation of the intent of the GPL to what the GPL actually says. It doesn't say anything about you having the right to modify source code and run it on the hardware that carried the binary version. It doesn't even address anything similar to that.

I have no problem with the fact that you don't like what TiVo did. But at least be intellectually honest about the issue. What you don't like is how their decision deviated from what YOU want, and how YOU define "free". It doesn't violate the GPL at all.

I can't believe how hard it is to get this point across!

You're having trouble making your point because you keep using leaps of logic, false allegations, and non-sequiturs.

You're right, of course...

Freeman's picture

...You're always right, aren't you? Same here; I thought I was wrong once, but I was mistaken! ;)

The error in your logic starts when you add your own personal interpretation of the intent of the GPL to what the GPL actually says. It doesn't say anything about you having the right to modify source code and run it on the hardware that carried the binary version. It doesn't even address anything similar to that.

If it's my "own personal interpretation of the intent of the GPL", then why is the FSF making it more explicit in GPLv3 by adding the requirement to "include[] any encryption or authorization keys necessary to install and/or execute modified versions from source code in the recommended or principal context of use, such that they can implement all the same functionality in the same range of circumstances."? I must have a hell of a lot of influence with them! This isn't just my interpretation, it's the FSF's. You know: the folks who hold the copyright to the GPL. I get these ideas from reading their work. You ought to try it sometime. The Linux kernel developers should too.

Not that I'm always right in the way I try to get my point across. I'll agree with you on the last two sentences I quoted above (in the context of GPLv2) and also with this:
You believe that free implies the ability to modify the code and run it on specific hardware. But you have yet to cite a single line of the GPL that says anything of the sort.

You're absolutely right. Wrong argument to make, nothing to back it up. Guilty as charged, and I thank you for graciously allowing me the opportunity to be the first to admit a mistake, because I think we've both made plenty.

Here's how I should be trying to get the point across: From the GPLv2 preamble:
"For example, if you distribute copies of such a program, whether gratis or for a fee, you must give the recipients all the rights that you have." From Section 6: "You may not impose any further restrictions on the recipients' exercise of the rights granted herein." The intent of the GPL is clear. Tivo is asserting the exclusive right to distribute GPL code that will install and execute on any Tivo while "imposing further restrictions" which prevent recipients of the GPL code that Tivo distributed to them from excersizing that right.

It's ironic, but I was reading RMS earlier this evening and he stated that Tivo could distribute the GPL code on ROM in the machine and be in compliance, since they would no longer be able to assert exclusive rights to distribute modifications to the code that anybody could install and execute in their Tivo. It is illustrative of the "equal rights" freedom that is the core purpose of the GPL: By contrast, the GNU General Public License is intended to guarantee your freedom to share and change free software--to make sure the software is free for all its users. (from GPLv2 preamble)

I think you've correctly identified the problem: "The error in your logic starts when you add your own personal interpretation of the intent of the GPL to what the GPL actually says", but have failed to recognize when those you are defending have done so. From the Kernel Developers' Position on GPLv3: We believe that the pre-eminent success of Linux owes a great part to the dynamism and diversity of its community of contributors, and that one of the catalysts for creating and maintaining this community is the development contract as expressed by GPLv2. They use the phrase "development (or developer) contract" twice more in the document in reference to GPLv2. But try as I might, I can't find any language in the GPLv2 that implies any sort of intent to serve as an expression of a development contract and in fact it explicitly states that it does not have the properties of a contract: 5. You are not required to accept this License, since you have not signed it. (GPLv2). If the Linux developers are referring to the requirement to distribute source code with Free Software, that is clearly declared in the GPL to be for the sole purpose of guaranteeing the end user's rights. From the preamble: When we speak of free software, we are referring to freedom, not price. Our General Public Licenses are designed to make sure that you have the freedom to distribute copies of free software (and charge for this service if you wish), that you receive source code or can get it if you want it, that you can change the software or use pieces of it in new free programs; and that you know you can do these things.

But I really must take exception to this: But at least be intellectually honest about the issue. What you don't like is how their decision deviated from what YOU want, and how YOU define "free".

I've clearly stated what it is I don't like about what Tivo did, and I've just cited the GPLv2 to back up my previous statements as to the purpose of the GPL, the FSF's definition of "free", and my case for declaring Tivo in violation of that purpose. If you were being intellectually honest you wouldn't be accusing the FSF of wanting to dictate hardware design or accusing those who object to Tivo's actions of shady ulterior motives. For your information, I would modify a GPLv3-compliant Tivo to do things like show slideshows and display other legal media stored on my home network and from the internet, run an asterisk client so I can see caller-id on the TV screen, etc., not that it's any of your or anyone else's business.

I think Glyn has really nailed it with the observation that the different perspectives are driving the conflict. I was initially appalled at your attitude towards Free Software when we began this discussion over on your article . As Editor in Chief of a magazine devoted to a popular Free Software project, I had assumed your views would be closer aligned to those of the FSF. But I've come to the slow realization that it's called Linux Journal, not Free Software Journal, and as such, your views are more properly aligned to those of the Linux kernel developers, who do have some legitimate issues with the FSF's proposed changes to the GPL right now, and I shouldn't be surprised to see you vigorously defending their viewpoint. It's your job, right? Perspective: I was looking at Linux Journal from the perspective of Free Software and the Linux kernel developers are looking at the GPL from an Open-Source perspective, and neither of us are able to clearly see the other viewpoint from those perspectives.

You're having trouble making your point because you keep using leaps of logic, false allegations, and non-sequiturs.

I'll let that one go. I've said worse about you and I wasn't any more in the right to do so. I apologize for that and I really do appreciate your participation with me in this discussion, as it's served to hone my understanding of the Tivoization issue and deepen my appreciation for the wisdom of the FSF and the GPL.

I respectfully disagree

Nicholas Petreley's picture

I don't think the FSF is trying to use Linux or GNU/Linux as a tool against these things

I think the FSF is doing just that. Otherwise the term Tivoization wouldn't exist.

I'm afraid I don't understand

Glyn Moody's picture

Could you please explain your thinking on that?

Read the dialog between the two sides

Nicholas Petreley's picture

Much of the focus of their discussion is on things like anti-Tivo clauses, tivoization, etc. If TiVo is such a big issue to both sides, what is the FSF focusing on if not Linux, then? TiVo, cell phones, etc., that use a GPL kernel aren't using emacs as the kernel. They're using Linux. An anti-Tivo clause would therefore have no effect whatsoever on TiVo and any company like it unless Linux adopts the GPLv3. Therefore if FSF wants the GPLv3 to have any realistic effect on the industry, the Linux developers have to adopt it.

Funny. I could swear you

Anonymous's picture

Funny. I could swear you used to remember that Linux (and for that matter, often Solaris) used the GNU tools. They're not just for development. I don't think it'd work without them.
One result of all this might be a true schism, in the sense that if the GNU people find themselves forced to push the HURD, the Linus people will be forced to develop GNUless Linux. Tough to say which is going to be a bigger deal. While it's true that HURD sucks, it's also true that more than one OS has run on the GNU tools, while Linux has never run on anything else.
Of course I suppose Linux could fork GNU from the moment before GNU went to version 3.

Perversely...

Glyn Moody's picture

..I don't think TiVo itself is a big issue for the FSF: what *is* a big issue is the broader danger it represents.

So yes, you're right, TiVo won't be affected if Linux doesn't adopt GPLv3, but the FSF would probably say that's an issue for the consciences of the kernel coders; at least FSF has provided a way to plug that hole. In a sense, the FSF would be failing in its duty to the community if it didn't come up with a way to thwart such uses - whether or not the community chooses to adopt that particular solution.

Ideally, I expect FSF would prefer people to use the Hurd for the kernel, which presumably will be released under GPLv3, and which would therefore thwart TiVo-type uses. Of course that won't happen, so it's completely theoretical. But as I've emphasised, for the purists, the theory is just as important as the practice when it comes to principles....

Theory/Practice

dhlii's picture

Stallman and the FSF are not about theory, they are about practice. But they also are not about numbers. If two people are running a truly free system, they have met their objective.

The principles of the FSF are very dear to them. The numbers of people using "Free" software are less important.

To the same extent they are not "purists" as much as they are principled. They are not dreaming about utopia's they are actively trying to create them. They beleive in the merit of their view point and they beleive that if they can instantiate and instance of it in practice that it will be naturally attractive and grow by virtue of its merits.

You can read Stallman or his detractors, and get a sense that he is some kind of radical loon. But everyone living in the Linux/Open Source world is the direct beneficiary of his efforts.

Moreover, the exisitance of an enormous body of software licensed under the GPLv2, speaks strongly to the fact that the ideas of the FSF are not just interesting radical utopian principles, but a real world functional working system.

How many utopian systems are you aware of that actually worked when put into practice ?

One of the most critical aspects of the whole Free/Open Source software debate, is that despite all the tirades and objections, the GPL actually works. Linus who still does nto actually grasp the significance and importance of the principles it stands for adopted it and the success of Linux is in no small part the result of that.

The GPL works for Open Source developers that do not quite believe in it.

The GPL works for businesses like IBM or even for Sun despite getting drug kicking and screaming into it.

Whether inspite of - or more likely because of its perceived flaws, the GPL works.

Despite rants to the contrary the GPL is inherently capitolistic. It encourages the development of software and ensures no one gets an unfair advantage.

No legislative body I am aware of could possibly create anything so elegant and effective

Hurd & BSD

Anonymous's picture

Ideally, I expect FSF would prefer people to use the Hurd for the kernel, which presumably will be released under GPLv3, and which would therefore thwart TiVo-type uses.

The Hurd has been in pre-production development for over 15 years. That's a long time in CS. I think one can be excused for dismissing it as a viable alternative to Linux.

But I've often wondered why GNU doesn't just fork a BSD kernel and relicense it under the GPL.

Talk about useless

Nicholas Petreley's picture

Of course that won't happen, so it's completely theoretical. But as I've emphasised, for the purists, the theory is just as important as the practice when it comes to principles....

That's just what we need. An impractical license that exists to make a political statement and will only be implemented in a kernel nobody will use. Well, if FSF wants to spin its wheels on something like that, I guess that's fine with me.

Give it up already

tecknogyk's picture

For the umpteenth time...

Would people like you give it up already!? Christ! Let me make this real clear for you. As the copyright holder I absolutely have the right to dictate the conditions for distribution of my software. This means that I can indeed dictate how a hardware manufacturer uses it. If they don't like those conditions then they don't use my software. It really is that simple. This is the same damn debate that we've had with BSD people. There's nothing to debate here. I as a developer can use whatever restrictions I want and you are free not to like it, but it's my choice. Don't like it, don't use my software. So let's stop with all the, "A software license has no business..." crap. A software license has every right to dictate the conditions that software can be distributed under. You may not like it, but that doesn't make it any less true. Another way of looking at it is that it's awfully arrogant of people to try and dictate what kind of restrictions I put on the use of my software. If a hardware company can build in restrictions for the use of their hardware then why can't I put in restrictions on the use of my software? What makes them better than me? The answer: absolutely nothing. They don't have to use my software if they don't like the conditions. End of story. Get over it people.

i feel your thought

Anonymous's picture

I feel your thought and as an OSS developer, i'm definitely going to use the GPL v3 for all my software. We see who wins the "Lock Out" game.

The implications of Tivoisation are almost immediate in the case of users of OSS but not so for developers of OSS. What happens when most commercial vendors ship crippled (DRM'ed) GPL v2 software?

Users cannot turn to the OSS developers to fix it (Vendor lock in again)
OSS developers cannot fix their own code for the users (Vendor lock out of your own code)
So only and only the vendors can fix the code for the users (Remind you of closed source software?).

With this, i could see M$ striking a deal, wait! don't intel and AMD chips ship with some kind of DRM stuff now. M$ doen't need to strike a deal, it could as well Tivoise all GPL v2 code and lock in every body. say .... imagine Windows server 2008 is Tivoised GPL v2 code + Linux that no OSS developer can fix.
Hmmmmm, I wonder where we are headed

If this is not addressed now, the implications will quickly turn from bad to very-very-bad.

Webinar
One Click, Universal Protection: Implementing Centralized Security Policies on Linux Systems

As Linux continues to play an ever increasing role in corporate data centers and institutions, ensuring the integrity and protection of these systems must be a priority. With 60% of the world's websites and an increasing share of organization's mission-critical workloads running on Linux, failing to stop malware and other advanced threats on Linux can increasingly impact an organization's reputation and bottom line.

Learn More

Sponsored by Bit9

Webinar
Linux Backup and Recovery Webinar

Most companies incorporate backup procedures for critical data, which can be restored quickly if a loss occurs. However, fewer companies are prepared for catastrophic system failures, in which they lose all data, the entire operating system, applications, settings, patches and more, reducing their system(s) to “bare metal.” After all, before data can be restored to a system, there must be a system to restore it to.

In this one hour webinar, learn how to enhance your existing backup strategies for better disaster recovery preparedness using Storix System Backup Administrator (SBAdmin), a highly flexible bare-metal recovery solution for UNIX and Linux systems.

Learn More

Sponsored by Storix